Add missing formatters, cli config support

src/cli/commands/check.ts

@@ -4,6 +4,7 @@ import { formatText } from '../formatters/text.js';
 import { formatJson } from '../formatters/json.js';
 import { formatScorecard } from '../formatters/scorecard.js';
 import type { SamplingStrategy } from '../../types.js';
+import { findConfig } from '../../helpers/config.js';
 
 // Ensure all checks are registered
 import '../../checks/index.js';
@@ -13,28 +14,48 @@ const FORMAT_OPTIONS = ['text', 'json', 'scorecard'] as const;
 
 export function registerCheckCommand(program: Command): void {
   program
-    .command('check <url>')
+    .command('check [url]')
     .description('Run agent-friendly documentation checks against a URL')
+    .option('--config <path>', 'Path to config file (default: auto-discover agent-docs.config.yml)')
     .option('-f, --format <format>', 'Output format: text, json, or scorecard', 'text')
     .option('-c, --checks <checks>', 'Comma-separated list of check IDs to run')
-    .option('--max-concurrency <n>', 'Maximum concurrent requests', '3')
-    .option('--request-delay <ms>', 'Delay between requests in ms', '200')
-    .option('--max-links <n>', 'Maximum links to test', '50')
-    .option(
-      '--sampling <strategy>',
-      'URL sampling strategy: random, deterministic, or none',
-      'random',
-    )
-    .option('--pass-threshold <n>', 'Pass threshold in characters', '50000')
-    .option('--fail-threshold <n>', 'Fail threshold in characters', '100000')
+    .option('--max-concurrency <n>', 'Maximum concurrent requests')
+    .option('--request-delay <ms>', 'Delay between requests in ms')
+    .option('--max-links <n>', 'Maximum links to test')
+    .option('--sampling <strategy>', 'URL sampling strategy: random, deterministic, or none')
+    .option('--pass-threshold <n>', 'Pass threshold in characters')
+    .option('--fail-threshold <n>', 'Fail threshold in characters')
     .option('-v, --verbose', 'Show per-page details for checks with issues')
     .option('--fixes', 'Show fix suggestions for warn/fail checks')
     .option('--score', 'Include scoring data in JSON output')
-    .action(async (rawUrl: string, opts: Record<string, string>) => {
-      const url = normalizeUrl(rawUrl);
-      const checkIds = opts.checks ? opts.checks.split(',').map((s) => s.trim()) : undefined;
-      const format = opts.format as string;
+    .action(async (rawUrl: string | undefined, opts: Record<string, unknown>) => {
+      // Load config: explicit path or auto-discover
+      let config;
+      try {
+        config = await findConfig(opts.config as string | undefined);
+      } catch (err) {
+        process.stderr.write(`Error: ${(err as Error).message}\n`);
+        process.exitCode = 1;
+        return;
+      }
 
+      // Resolve URL: CLI arg > config url > error
+      const resolvedUrl = rawUrl ?? config?.url;
+      if (!resolvedUrl) {
+        process.stderr.write(
+          'Error: No URL provided. Pass a URL as an argument or set "url" in agent-docs.config.yml\n',
+        );
+        process.exitCode = 1;
+        return;
+      }
+      const url = normalizeUrl(resolvedUrl);
+
+      // Resolve options: CLI flags > config > hardcoded defaults
+      const checkIds = opts.checks
+        ? (opts.checks as string).split(',').map((s) => s.trim())
+        : config?.checks;
+
+      const format = opts.format as string;
       if (!FORMAT_OPTIONS.includes(format as (typeof FORMAT_OPTIONS)[number])) {
         process.stderr.write(
           `Error: Invalid format "${format}". Must be one of: ${FORMAT_OPTIONS.join(', ')}\n`,
@@ -43,7 +64,9 @@ export function registerCheckCommand(program: Command): void {
         return;
       }
 
-      const sampling = opts.sampling as SamplingStrategy;
+      const samplingRaw =
+        (opts.sampling as string | undefined) ?? config?.options?.samplingStrategy ?? 'random';
+      const sampling = samplingRaw as SamplingStrategy;
       if (!SAMPLING_STRATEGIES.includes(sampling)) {
         process.stderr.write(
           `Error: Invalid sampling strategy "${sampling}". Must be one of: ${SAMPLING_STRATEGIES.join(', ')}\n`,
@@ -52,6 +75,31 @@ export function registerCheckCommand(program: Command): void {
         return;
       }
 
+      const maxConcurrency = parseInt(
+        String((opts.maxConcurrency as string | undefined) ?? config?.options?.maxConcurrency ?? 3),
+        10,
+      );
+      const requestDelay = parseInt(
+        String((opts.requestDelay as string | undefined) ?? config?.options?.requestDelay ?? 200),
+        10,
+      );
+      const maxLinksToTest = parseInt(
+        String((opts.maxLinks as string | undefined) ?? config?.options?.maxLinksToTest ?? 50),
+        10,
+      );
+      const passThreshold = parseInt(
+        String(
+          (opts.passThreshold as string | undefined) ?? config?.options?.thresholds?.pass ?? 50000,
+        ),
+        10,
+      );
+      const failThreshold = parseInt(
+        String(
+          (opts.failThreshold as string | undefined) ?? config?.options?.thresholds?.fail ?? 100000,
+        ),
+        10,
+      );
+
       if (format !== 'json') {
         const parsed = new URL(url);
         const target =
@@ -63,13 +111,13 @@ export function registerCheckCommand(program: Command): void {
 
       const report = await runChecks(url, {
         checkIds,
-        maxConcurrency: parseInt(opts.maxConcurrency, 10),
-        requestDelay: parseInt(opts.requestDelay, 10),
-        maxLinksToTest: parseInt(opts.maxLinks, 10),
+        maxConcurrency,
+        requestDelay,
+        maxLinksToTest,
         samplingStrategy: sampling,
         thresholds: {
-          pass: parseInt(opts.passThreshold, 10),
-          fail: parseInt(opts.failThreshold, 10),
+          pass: passThreshold,
+          fail: failThreshold,
         },
       });
 

src/cli/formatters/text.ts

@@ -123,6 +123,203 @@ const DETAIL_FORMATTERS: Record<string, DetailFormatter> = {
       return formatDetailLine('fail', b.url, info);
     });
   },
+
+  'rendering-strategy': (details) => {
+    const pages = details.pageResults as
+      | Array<{
+          url: string;
+          status: string;
+          analysis?: { spaMarker?: string | null; visibleTextLength?: number };
+          error?: string;
+        }>
+      | undefined;
+    if (!pages) return [];
+    return pages
+      .filter((p) => p.status !== 'pass')
+      .map((p) => {
+        if (p.error) return formatDetailLine('fail', p.url, p.error);
+        const marker = p.analysis?.spaMarker;
+        const textLen = p.analysis?.visibleTextLength ?? 0;
+        const info = marker
+          ? `SPA shell (${marker}, ${textLen} chars visible)`
+          : `sparse content (${textLen} chars visible)`;
+        return formatDetailLine(p.status, p.url, info);
+      });
+  },
+
+  'redirect-behavior': (details) => {
+    const pages = details.pageResults as
+      | Array<{ url: string; classification: string; redirectTarget?: string; error?: string }>
+      | undefined;
+    if (!pages) return [];
+    return pages
+      .filter(
+        (p) =>
+          p.classification === 'cross-host' ||
+          p.classification === 'js-redirect' ||
+          p.classification === 'fetch-error',
+      )
+      .map((p) => {
+        if (p.error) return formatDetailLine('fail', p.url, p.error);
+        const target = p.redirectTarget ? ` → ${p.redirectTarget}` : '';
+        return formatDetailLine(
+          p.classification === 'cross-host' ? 'warn' : 'fail',
+          p.url,
+          `${p.classification}${target}`,
+        );
+      });
+  },
+
+  'auth-gate-detection': (details) => {
+    const pages = details.pageResults as
+      | Array<{
+          url: string;
+          classification: string;
+          status?: number | null;
+          hint?: string;
+          ssoDomain?: string;
+          error?: string;
+        }>
+      | undefined;
+    if (!pages) return [];
+    return pages
+      .filter((p) => p.classification !== 'accessible')
+      .map((p) => {
+        if (p.error) return formatDetailLine('fail', p.url, p.error);
+        let info = p.classification;
+        if (p.ssoDomain) info += ` (${p.ssoDomain})`;
+        else if (p.hint) info += ` (${p.hint})`;
+        else if (p.status) info += ` (HTTP ${p.status})`;
+        return formatDetailLine('fail', p.url, info);
+      });
+  },
+
+  'llms-txt-directive': (details) => {
+    const pages = details.pageResults as
+      | Array<{ url: string; found: boolean; positionPercent?: number; error?: string }>
+      | undefined;
+    if (!pages) return [];
+    return pages
+      .filter((p) => !p.found || (p.positionPercent != null && p.positionPercent > 10))
+      .map((p) => {
+        if (p.error) return formatDetailLine('fail', p.url, p.error);
+        if (!p.found) return formatDetailLine('fail', p.url, 'no directive found');
+        return formatDetailLine('warn', p.url, `directive at ${p.positionPercent}% of page`);
+      });
+  },
+
+  'tabbed-content-serialization': (details) => {
+    const pages = details.tabbedPages as
+      | Array<{
+          url: string;
+          status: string;
+          tabGroups?: unknown[];
+          totalTabbedChars?: number;
+          error?: string;
+        }>
+      | undefined;
+    if (!pages) return [];
+    return pages
+      .filter((p) => p.status !== 'pass')
+      .map((p) => {
+        if (p.error) return formatDetailLine('fail', p.url, p.error);
+        const groups = p.tabGroups?.length ?? 0;
+        const size = formatSize(p.totalTabbedChars ?? 0);
+        return formatDetailLine(p.status, p.url, `${groups} tab groups, ${size} serialized`);
+      });
+  },
+
+  'markdown-content-parity': (details) => {
+    const pages = details.pageResults as
+      | Array<{
+          url: string;
+          status: string;
+          missingPercent?: number;
+          sampleDiffs?: string[];
+          error?: string;
+        }>
+      | undefined;
+    if (!pages) return [];
+    return pages
+      .filter((p) => p.status !== 'pass')
+      .map((p) => {
+        if (p.error) return formatDetailLine('fail', p.url, p.error);
+        const pct =
+          p.missingPercent != null ? `${Math.round(p.missingPercent)}% missing` : 'content differs';
+        return formatDetailLine(p.status, p.url, pct);
+      });
+  },
+
+  'http-status-codes': (details) => {
+    const pages = details.pageResults as
+      | Array<{
+          url: string;
+          testUrl?: string;
+          classification: string;
+          status?: number | null;
+          bodyHint?: string;
+          error?: string;
+        }>
+      | undefined;
+    if (!pages) return [];
+    return pages
+      .filter((p) => p.classification !== 'correct-error')
+      .map((p) => {
+        if (p.error) return formatDetailLine('fail', p.testUrl ?? p.url, p.error);
+        const info = p.bodyHint
+          ? `HTTP ${p.status} (${p.bodyHint})`
+          : `HTTP ${p.status} instead of 404`;
+        return formatDetailLine('fail', p.testUrl ?? p.url, info);
+      });
+  },
+
+  'cache-header-hygiene': (details) => {
+    const endpoints = details.endpointResults as
+      | Array<{
+          url: string;
+          status: string;
+          effectiveMaxAge?: number | null;
+          noStore?: boolean;
+          error?: string;
+        }>
+      | undefined;
+    if (!endpoints) return [];
+    return endpoints
+      .filter((e) => e.status !== 'pass')
+      .map((e) => {
+        if (e.error) return formatDetailLine('fail', e.url, e.error);
+        if (e.noStore) return formatDetailLine(e.status, e.url, 'no-store');
+        if (e.effectiveMaxAge == null) return formatDetailLine(e.status, e.url, 'no cache headers');
+        const age = e.effectiveMaxAge;
+        const human =
+          age >= 86400
+            ? `${Math.round(age / 86400)}d`
+            : age >= 3600
+              ? `${Math.round(age / 3600)}h`
+              : `${age}s`;
+        return formatDetailLine(e.status, e.url, `max-age ${human}`);
+      });
+  },
+
+  'section-header-quality': (details) => {
+    const analyses = details.analyses as
+      | Array<{
+          url: string;
+          framework?: string;
+          genericHeaders?: number;
+          totalHeaders?: number;
+          hasGenericMajority?: boolean;
+        }>
+      | undefined;
+    if (!analyses) return [];
+    return analyses
+      .filter((a) => a.hasGenericMajority)
+      .map((a) => {
+        const ratio = `${a.genericHeaders}/${a.totalHeaders} generic`;
+        const fw = a.framework ? ` (${a.framework})` : '';
+        return formatDetailLine('warn', a.url, `${ratio}${fw}`);
+      });
+  },
 };
 
 function formatDetailLine(status: string, url: string, metric: string): string {

src/helpers/config.ts

@@ -38,3 +38,40 @@ export async function loadConfig(dir?: string): Promise<AgentDocsConfig> {
     `No agent-docs config file found. Create ${CONFIG_FILENAMES[0]} with at least a "url" field.`,
   );
 }
+
+/**
+ * CLI-oriented config loader:
+ * - If explicitPath is given, reads that file directly and throws if not found.
+ * - Otherwise, auto-discovers by walking up from startDir (default: cwd).
+ * - Returns null if no config file is found (instead of throwing).
+ * - Does not require the "url" field — the CLI can supply it via argument.
+ */
+export async function findConfig(
+  explicitPath?: string,
+  startDir?: string,
+): Promise<AgentDocsConfig | null> {
+  if (explicitPath) {
+    const filepath = resolve(process.cwd(), explicitPath);
+    const content = await readFile(filepath, 'utf-8');
+    return parseYaml(content) as AgentDocsConfig;
+  }
+
+  let searchDir = resolve(startDir ?? process.cwd());
+  while (true) {
+    for (const filename of CONFIG_FILENAMES) {
+      const filepath = resolve(searchDir, filename);
+      try {
+        const content = await readFile(filepath, 'utf-8');
+        return parseYaml(content) as AgentDocsConfig;
+      } catch (err) {
+        if ((err as NodeJS.ErrnoException).code === 'ENOENT') continue;
+        throw err;
+      }
+    }
+    const parent = dirname(searchDir);
+    if (parent === searchDir) break;
+    searchDir = parent;
+  }
+
+  return null;
+}

src/helpers/index.ts

@@ -1,4 +1,4 @@
-export { loadConfig } from './config.js';
+export { loadConfig, findConfig } from './config.js';
 export { describeAgentDocs, describeAgentDocsPerCheck } from './vitest-runner.js';
 export { looksLikeMarkdown, looksLikeHtml } from './detect-markdown.js';
 export {